Vitamin C prevents colds

Taking vitamin C every day does not stop most people from catching colds. The largest pooled analysis of the question combined 29 trial comparisons and 11,306 participants and found a risk ratio of 0.97 in general community trials, with a confidence interval running from 0.94 to 1.00. That is a 3% reduction in the proportion of people who caught at least one cold over the study period, with an interval that reaches 1.00, the point of no effect. Trials using less than 200 mg a day were excluded, as were trials without a placebo comparison, so this is not a result about token doses. A 2025 review that pooled 22 adult trials, including those published since, found a similarly small lean at 500 to 1,000 mg a day (a risk ratio of 0.93, interval 0.87 to 1.00) that did not hold once trials in soldiers and athletes were set aside.

There is one exception and it is narrow. In five trials covering 598 marathon runners, skiers and soldiers on subarctic exercises, the same review found a risk ratio of 0.48, close to halving the number of people who caught a cold. All five involved a brief period of heavy physical effort. The same review lists two trials of longer physical stress, in 674 US Marine recruits over two months (2,000 mg a day) and in 39 competitive young swimmers over three months, and neither found a difference in colds. A later trial in 1,444 Republic of Korea Army recruits, taking 6,000 mg a day through a 30-day basic training period, reported an odds ratio of 0.80 for a recorded cold diagnosis in the vitamin C group. An odds ratio is a different measure from a risk ratio. Counted as plain proportions, a comment on the journal's page reads the trial's first table as 53.8% of recruits taking vitamin C and 58.7% of those taking placebo having a recorded cold (P = 0.059).

Two neighbouring questions have their own answers. The same review found that regular daily supplementation shortened colds that did happen, by about 8% in adults and 14% in children, and reduced their severity. It also found no consistent effect when vitamin C was started after symptoms had already begun. Those are the duration question and the therapy question, and they are covered as separate claims rather than folded in here, because they are different questions with different answers.

Dose is a smaller lever than the labels suggest. Plasma vitamin C is tightly regulated and approaches saturation somewhere around 200 to 400 mg a day, with the absorbed fraction falling and renal excretion rising as intake climbs. Most trials in the pooled analysis used 1 g a day or more, which sits at or past that plateau. The trials therefore largely compared people held at plasma saturation with people somewhat below it, rather than comparing a large dose against none.

Starting status appears to matter more than dose. The NIH Office of Dietary Supplements describes the same overall picture and adds that regular intake of at least 200 mg a day might help people with marginal vitamin C status, naming older adults and people who regularly smoke alongside those exposed to extreme physical exercise or cold environments. Only one trial has tested that group directly: 28 young men with low plasma vitamin C, in which 7 of 15 taking 1,000 mg a day and 11 of 13 taking placebo reported at least one cold over eight winter weeks. It was small and was not registered in advance, and the total number of colds did not differ significantly between the groups, so it points in a direction rather than settling the question.

The biology is mapped in unusual detail. Vitamin C accumulates in neutrophils and other phagocytic cells, supports the epithelial barrier against pathogens, and contributes to chemotaxis, phagocytosis and microbial killing. Deficiency produces measurably impaired immunity and greater susceptibility to infection. A mapped pathway and a measured outcome are different things, and here they do not line up: the mechanism is clear and the effect on cold incidence in people who are already replete is close to zero.

On how current this is. The question has been studied for roughly 70 years and the Cochrane review has run through four versions, in 2000, 2004, 2007 and 2013, each covering a larger trial set. Its answer on incidence did not move across any of them. The 2025 review searched the literature through February 2025, picked up the trials published after the last Cochrane update, and did not move it either. A replacement Cochrane review covering the broader upper respiratory question was registered as a protocol in March 2019, and a literature search on September 28, 2026 returns that protocol and no completed review under it.